Bihar’s big question: Did counted votes exceed polled votes?

A seemingly striking discrepancy at the heart of Bihar’s recent Assembly election statistics has led political economist Parakala Prabhakar to publicly confront the Election Commission of India (ECI): how did the number of votes counted in the 2025 Bihar Assembly polls exceed the number of votes that the ECI’s own turnout figures indicate were polled?

In a detailed 12-part thread on X, Prabhakar lays out what he calls a series of numerical red flags — gaps, revisions and omissions — that, he argues, warrant immediate clarification from the Commission.

Prabhakar’s central concern is straightforward. Using the ECI’s final electorate figure — 7,45,26,858 electors, published after polling — and the Commission’s declared voter turnout of 67.13 per cent, he calculates that Bihar should have recorded roughly 5,00,29,880 votes polled.

Yet when he adds up the votes counted across all 243 Assembly constituencies, the total is 5,02,07,553. That difference of 1,77,673 forms the core of his demand: a transparent explanation of how counted votes came to exceed the ECI’s own estimate of polled votes.

This numerical gap, he argues, is inseparable from a second pattern — the ECI’s multiple revisions of Bihar’s electorate in the weeks between the completion of the Special Intensive Revision (SIR) and the close of polls.

On 30 September, after announcing the SIR “successfully completed”, the Commission put the electorate at “~7.42 crore” (where ~ stands for approximately). Prabhakar questions this approximation, saying an SIR is meant to produce a precise figure.

On 6 October, the ECI updated the total to 7,43,55,976 after adding service voters. Then on 11 November, post-poll, the figure shifted again to 7,45,26,858. These moving baselines, he argues, complicate scrutiny of turnout and inflate uncertainty around what should be fixed numbers.

A third plank of his critique concerns data disclosure. While the ECI routinely releases turnout percentages, Prabhakar notes that it does not publish the straightforward, segment-wise gross number of votes polled alongside them. Without that basic aggregate, he argues, independent verification becomes impossible — and gaps such as the one he highlights cannot be cross-checked against official polled-vote tallies.

Prabhakar’s interrogation of Bihar’s numbers sits within a broader pattern. During the Maharashtra Assembly elections last year, he flagged what he described as an unprecedented late jump in turnout — a surge he estimated at over 76 lakh additional votes appearing after the close of polling.

He argued then that such a post-poll revision was historically abnormal and should have been accompanied by a segment-wise explanation of how the new figures were derived. The absence of such a reconciliation, he says, mirrors the opacity he now sees in Bihar.

In the present case, the ECI has not issued a point-by-point public reply to Prabhakar’s calculations or to the specific inconsistencies he has identified, though in the wider controversy it has dismissed allegations of manipulation and advised complainants to file objections through the statutory procedures under the Registration of Electors Rules.

For Prabhakar, however, the core issue is not motive but mathematics: unless the Commission publishes a single reconciled dataset covering electors, gross votes polled and final counts, he argues, the integrity of the process will remain clouded by unanswered questions.